The clinical utility of the use of more accurate dose calculation algorithms for radiotherapy treatment planning has not yet been shown for many clinical sites. Although increasingly sophisticated and accurate dose calculation methods such as Monte Carlo and convolution/superposition have over the past decade become increasingly available for clinical study, direct study of the usefulness of the more accurate dose distributions obtained with these algorithms in terms of improving clinical patient outcomes has not been performed. In this study, we will use the clinical outcomes from two conformal therapy treatment studies (parotid-sparing in the head/neck and dose escalation for non-small cell lung cancer) to investigate whether dose distributions calculated with the more accurate Monte Carlo method will improve the correlation between the clinical patient outcomes and the calculated dose distributions. Since both clinical complications and local control for the two clinical studies are well-documented, and the full 3-D patient anatomy and treatment information is available for the patients on those protocols, these data provide a unique opportunity to evaluate the potential for clinical improvements due to improved dose calculations with a retrospective study. To accomplish the proposed goal, we must: (a) perform the necessary algorithmic verification against measurements made in phantoms that closely resemble the relevant clinical geometries, and (b) use the validated and accurate calculational method to re-evaluate the dose distributions delivered to patients treated on the conformal therapy trials performed in the head/neck and lung, and determine if the clinical outcomes (complications and local control) are better correlated with the more accurate calculational results than with the original results. [unreadable] [unreadable]